Using FME To Efficiently Create NationalScale Vector Contours

  • Slides: 25
Download presentation
Using FME To Efficiently Create National-Scale Vector Contours From High. Resolution DEM Data Todd

Using FME To Efficiently Create National-Scale Vector Contours From High. Resolution DEM Data Todd Lewis President, Terra. Logik. com

Outline § § § Intro Problem Description Technical Issues Solution Approach w/ Demo Workspaces

Outline § § § Intro Problem Description Technical Issues Solution Approach w/ Demo Workspaces Q&A 2

Introduction § NAV CANADA produces navigation charts for general aviation (Visual Flight Rules Navigation

Introduction § NAV CANADA produces navigation charts for general aviation (Visual Flight Rules Navigation Charts – VNC) § Charts are used in visual conditions (3 -5 nautical miles visibility) § Charts are based upon standards set by the UN International Civil Aviation Org (ICAO) in Montreal § Terra. Logik designed the processes and procedures for topographic data supporting the VNC 3

Hurry up Jake! Turn the prop and get in! The owners are coming!!! 4

Hurry up Jake! Turn the prop and get in! The owners are coming!!! 4

Problem Definition § VFR Navigation Charts series composed of 52 charts covering the Canadian

Problem Definition § VFR Navigation Charts series composed of 52 charts covering the Canadian Territory, and 7 Terminal Area Charts. § VNC series contains 70+ layers of topographic information from multiple sources. § ICAO specifications provide guidelines on how to depict each layer of information. § Application of ICAO guidelines is inconsistent across the entire VNC series. 5

Problem Definition (2) § Fitness for Purpose is a key consideration: § Lambert Conformal

Problem Definition (2) § Fitness for Purpose is a key consideration: § Lambert Conformal Projection to enable accurate greatcircle distance/bearing calculations § Obstacle and terrain avoidance § Cartographic considerations such as labeling, feature context, scale, etc. § Visual reference for navigation in-flight for ground -based features, primarily: § § § Water Roads/rail Utilities Landforms Built-up areas and remote buildings/structures 6

Our Approach § Standardize the depiction and cartographic rules to consistently present VNC terrain

Our Approach § Standardize the depiction and cartographic rules to consistently present VNC terrain data across the entire chart series. § Develop automated processes to do the heavy lifting in application of cartographic rules to create a chart. § Document procedures and workflows to enable production operators to perform their work within an ISO 9002 environment. 7

Deliverable : Terrain Data File § Produce a Terrain Data File with agreed-to: §

Deliverable : Terrain Data File § Produce a Terrain Data File with agreed-to: § Format (# files, depictions, settings, etc. ) § Structure (level names, order, content/level) § Content (level of detail to show, rules) § Document procedures and workflows for production of charts using ETL and CAD tools § QA/QC checklists and manual editing procedures to clean-up Terrain Data File before passing to aeronautical information integration and prepress finishing procedures 8

Terrain Data File

Terrain Data File

Topographic Data File

Topographic Data File

Labels Data File

Labels Data File

Hypsometry Data File

Hypsometry Data File

Shaded Relief Data File

Shaded Relief Data File

Elevation Technical Issues § Contour and hypsometry required at 500 -ft intervals § Elevation

Elevation Technical Issues § Contour and hypsometry required at 500 -ft intervals § Elevation data not consistently available at 1: 50, 000 -scale across Canada § Integration with US data at border areas § US data available at 1” below 49, and 2” in Alaska § Source data is CDED 1: 50 K, 1: 250 K, ASTER GDEM, and USGS DEM 15

Red is missing CDED data. Elevation data, where missing, will be populated using NASA

Red is missing CDED data. Elevation data, where missing, will be populated using NASA ASTER data. Charts Affected by Elevation Issues: Alert Hazen Strait ere E m lles • Alert • Ellesmere • Devon Island • Baffin Bay • Cumberland Peninsula • Frobisher Bay • Ungava • Foxe Basin • Inukjuak • Coats Island • Baker Lake • Boothia • Resolute • Banks Island • Hazen Strait • Mackenzie Delta • Amundsen Gulf • Cambridge Bay • Klondike • Great Bear Lake • Bathurst Inlet • Rankin Inlet • Yellowknife • Fort Simpson • Kitimat • Anticosti a sk la A n Ban Devo Ma ks I Resolute d s ck Islan e De nz Am Bay und lta ie n i f sen n af Gulf rla la ia B h e t Kl o C a o mbridge B b su on um nin dik G B C a y W e rea e e hi t Be Fox in d P te a r Bathurst Lak as h r e aker Lk B e ors B he s Inlet i b Fo Fro Bay Sim rt pso Y At a n ellowkn s Is. lin av ife Rankin Coat g y Un Inlet For Ba k e a t s ju Ne lso Lake oo i Inuk W G ol n la st st Kit on udson Athab o h H im c asca i us Lake at nt ab Bay r Prin A W es de m a n Geo ce J ut rge Ed Ga Big Tro i Bay mont m i on Flin Flon Lake Va ut nco ico h n uv to ins. C er Cal c m r gar on hunde Tim y l. M Regina Winnipeg T Bay a e ntr o e i M Sault Ste. Mar o ont Tor y w H 16

Data Mismatch Canadian Elevation Data Missing Data CAN/US Border USGS Elevation Data 17

Data Mismatch Canadian Elevation Data Missing Data CAN/US Border USGS Elevation Data 17

Software Limitations § Each VNC Chart is composed of 350 -400 1: 50, 000

Software Limitations § Each VNC Chart is composed of 350 -400 1: 50, 000 CDED tiles § GIS & RS tools could create contour lines, but determining hypsometric tint requires polygon fills § Chart borders present issues with polygonizing contours § Volume of data presents huge challenges for processing and visualization 18

Requirements § Need to create: § Hypsometric Polygons at 500’, 1000’, 2000’, 3000’, 5000’,

Requirements § Need to create: § Hypsometric Polygons at 500’, 1000’, 2000’, 3000’, 5000’, 7000’, 9000’, 12000’ § Contours @ 500’ intervals <= 4000’ § Contours @ 1000’ intervals > 4000’ § 350 -400 1: 50, 000 tiles per chart § Contours, hypsometry, spot elevations, and the shaded relief need to align 19

Solution Ingredients § § § § Transformer Loop Raster. Expression. Evaluator Transformer GDAL for

Solution Ingredients § § § § Transformer Loop Raster. Expression. Evaluator Transformer GDAL for raster-to-vector conversion Python. Caller for coding GDAL Workspace. Runner Generalizer Throw in some algorithms to: § Select significant contours for print scale: § Area/Perimeter > Tolerance 20

Process: Tiles First § Break raster into one band per contour interval (1 -18

Process: Tiles First § Break raster into one band per contour interval (1 -18 bands total per tile) § Batch process the above with Workspace. Runner Synchronous/Async: § Time w/o Async: 20 hours § Time w/ Async: 1. 5 hours § Convert each band into a vector polygon layer § Use Python. Caller and GDAL to perform raster-to-vector conversion 21

Process: Merge & Generalize § Dissolve all polygons grouped by elevation level (500, 1000,

Process: Merge & Generalize § Dissolve all polygons grouped by elevation level (500, 1000, 1500, …) § But need to look for depressions, Donut. Hole. Extractor § But need to densify for generalization § Generalize § Look for generalization errors, polygons w/ Na. N for area are errors to be logged § Evaluate for print scale using algorithm: § Area/Perimeter > Tolerance (0. 0005 or 0. 001) 22

Process: Symbolize Select Hypsometry values Remove depressions from hypsometric levels Convert all polygons to

Process: Symbolize Select Hypsometry values Remove depressions from hypsometric levels Convert all polygons to contour lines Change depression contours to normal contours if on hypsometry level § Output hypsometry and contours to Microstation levels with symbology from seed file § § 23

Example Output (false color)

Example Output (false color)

Questions? Todd Lewis President Terra. Logik Information Systems Inc. T: 613. 862. 0999 E:

Questions? Todd Lewis President Terra. Logik Information Systems Inc. T: 613. 862. 0999 E: todd. lewis@terralogik. com